- Smart Agriculture and AI
- IoT and Edge/Fog Computing
- UAV Applications and Optimization
- Energy Efficient Wireless Sensor Networks
- Network Security and Intrusion Detection
- Mobile Ad Hoc Networks
- Spectroscopy and Chemometric Analyses
- Blockchain Technology Applications and Security
- Autonomous Vehicle Technology and Safety
- Fault Detection and Control Systems
- Formal Methods in Verification
- Distributed Control Multi-Agent Systems
- Model-Driven Software Engineering Techniques
- Adaptive Control of Nonlinear Systems
- Face recognition and analysis
- Privacy-Preserving Technologies in Data
- Security in Wireless Sensor Networks
- Vehicle emissions and performance
- Advanced Control Systems Optimization
- Photovoltaic System Optimization Techniques
- Human Mobility and Location-Based Analysis
- Evacuation and Crowd Dynamics
- Traffic Prediction and Management Techniques
- AI in cancer detection
- Robotic Path Planning Algorithms
KPR Institute of Engineering and Technology
2014-2025
M. Kumarasamy College of Engineering
2024
Konkuk University
2008-2023
Botswana International University of Science and Technology
2023
SRM Institute of Science and Technology
2023
Saveetha University
2023
University of Abou Bekr Belkaïd
2023
Deep learning (DL) is an exciting field of interest for many researchers and business. Due to a massive leap in DL based research, domains like Business, science government sectors make use various applications. This work puts forward the importance its application few critical electrical segments. Initially, introduction Artificial Intelligence (AI) Machine Learning (ML) presented. Then need popular architectures, algorithms frameworks used are A summary different techniques outlined,...
The aggressive waves of ongoing world-wide virus pandemics urge us to conduct further studies on the performability local computing infrastructures at hospitals/medical centers provide a high level assurance and trustworthiness medical services treatment patients, help diminish burden chaos management operations. Previous contributed tremendous progress dependability quantification existing paradigms (e.g., cloud, grid computing) remote data centers, while few works investigated performance...
The progression of mobile technologies from first generation (1G) to fifth (5G) networks has brought about noteworthy advancement. present network environment is characterized by the coexistence various and technologies, leading overlapping or non-overlapping coverage areas, creating a next-generation heterogeneous landscape. Handoffs within same (i.e., GSM vice versa) are known as horizontal homogeneous handoffs. Conversely, when handoffs entail transition different type, such moving CDMA...
Decalcification is crucial in enhancing the diagnostic accuracy and interpretability of cardiac CT images, particularly cardiovascular imaging. Calcification coronary arteries structures can significantly impact quality images hinder precise diagnostics. This study introduces a novel approach, Hybrid Models for Decalcify Cardiac (HMDC), aimed at clarity through effective decalcification. critical medical imaging, especially scans, where calcification accurate The proposed HMDC leverages...
The COVID-19 pandemic has infected tens of millions individuals around the world, and it is currently posing a worldwide health calamity. Wearing face mask in public places one most effective protection strategies, according to World Health Organization (WHO). Moreover, their effectiveness declined due incorrect use mask. In this scenario, recognition systems are anticipated ensure that people's faces covered with masks locations. Many people do not correctly wear inadequate practices,...
Smart healthcare using the cloud and Internet of Things (IoT) allows for remote patient monitoring, real-time data collection, improved security, cost-effective storage analysis data. This paper proposes an information-centric dissemination scheme (ICDS) smart services in cities. The proposed addresses time sensitiveness aims to ensure consistent dissemination. ICDS uses decision-tree learning classify requests based on time-sensitive features, allowing prioritization access. also involves...
Automated Vehicles (AVs) reform the automotive industry by enabling real-time and efficient data exchange between vehicles. While connectivity automation of vehicles deliver a slew benefits, they may also introduce new safety, security, privacy risks. Further, AVs rely entirely on sensor from other too. On hand, is susceptible to anomalies caused cyber-attacks, errors, faults, resulting in accidents fatalities. Hence, it essential create techniques for detecting identifying their sources...
The proliferation of unmanned aerial vehicles (UAVs) has spawned a variety intelligent services, where efficient coordination plays significant role in increasing the effectiveness cooperative execution. However, due to limited operational time and range UAVs, achieving highly coordinated actions is difficult, particularly unknown dynamic environments. This paper proposes multiagent deep reinforcement learning (MADRL)-based fusion-multiactor-attention-critic (F-MAAC) model for multiple UAVs’...
In recent years, container-based virtualization technology for edge and cloud computing has advanced dramatically. Virtualization solutions based on Docker Containers provide a more lightweight efficient virtual environment Edge cloud-based applications. Because their use is growing its own still in early phases, these technologies will face slew of security issues. Vulnerabilities malware container images are two serious concerns. The risk privilege escalation increased because containers...
Plant pest detection is an essential for food security, quality of life and steady agricultural economy. Through rapid infestation by pests insects various problems faced in enormous yield every year. Many researches are being carried out worldwide to identify scientific methodologies pests. In the recent past, several approaches based on automation image processing has come light way address this issue. paper we present a clustering technique, popular algorithm kernel-based fuzzy c-means...
In this paper additive fault detection and isolation method coupled with tolerant control architecture are developed in order to deal component faults for a rotorcraft based unmanned aerial vehicle (RUAV). The failure considered is malfunction internal components of the helicopter which occurs during maneuvers: rotor angular rate variations, etc. These lead from trivial catastrophic damage system. proposed reconfiguration on parameter estimation approach drives reconfigurable system (RCS)...
The need of communication protocols in todaypsilas environment increases as much the network explores. Many new kinds protocols, e.g. for information sharing, security, etc., are being developed day-to-day which often leads to rapid, premature developments. have not scaled satisfy important properties like deadlock and livelock freedom, since MDA focuses on rapid development rather than quality models. In order fix above, we introduce a 2-Phase strategy based UML state machine sequence...
The state of mind gets expressed via Emojis' and Text Messages for the huge population. Microblogging social networking sites emerged as a popular communication channels among internet users. Supervised text classifiers are used sentimental analysis in both general specific emotions detection with more accuracy. main objective is to include intensity predicting different texts formats from twitter, by considering context associated emoticons punctuations. novel Future Prediction Architecture...
In precision poultry farming, detecting multiple objects is challenging. While Convolutional Neural Networks (CNNs) excel in single-object detection, addressing the complexity of multi-object requires advanced approaches. To overcome this challenge, we implement algorithms like YOLOv8, SSD, and Faster RCNN. The primary goal to analyze their performance, focusing on accuracy, speed, adaptability. We aim balance computational efficiency with optimal resource utilization, considering hardware...
Over the past decades, substantial research has been undertaken in design of intelligent architecture for rotorcraft-based unmanned aerial vehicles (RUAV). Designing is a challenging problem because future RUAVs are utterly autonomous and their performance comparable with that manned vehicles. This paper deals development layered architectural framework addresses issue arising control systems. The consists two layers. high-level layer occupied by planning routines. In this level, waypoints...
Autonomous landing is a challenging and important task for rotorcraft-based unmanned aerial vehicle (RUAV). This paper presents an autonomous algorithm RUAV. The accurate achieved by generating reference trajectory using near-minimum-time maneuver method tracked with Lyapunov stabilizing approach. Along this the testing environment RUAV developed to validate our algorithm. Complete setup implemented run under QNX RTOS, based on PC104 embedded board. tested evaluated HILS (Hardware in loop...
RUAV (Rotorcraft based Unmanned Aerial Vehicle) have been exploited in various fields such as surveillance, reconnaissance and search. Most of the study this area is focused on single RUAV; however using multiple unmanned vehicles big advantage to accomplish mission a short time effective way. Moreover little amount research has undertaken development control systems. This paper proposes behavior-based decentralized approach that allows an carry out its own flying specified region while...